All the ED patients positive to or suspected of COVID-19 infection will be included in the study. Descriptive reports of the collected data will be produced regularly. Different levels of aggregation will be considered: single center, province, region. In addition to providing useful elements to describe the evolution of the epidemic, the results presented in these reports will provide an accurate clinical description of a large sample of COVID-19 patients.
The study data will be leveraged to identify the most important risk factors and clinical conditions to predict poor patient outcomes. The most relevant factors will be selected using state-of-the-art techniques.
These factors will be used as predictors in multivariable predictive models, which will be developed to estimate the expected prognosis of the patients. The study outcomes will be death or need of intubation within 7 days since ED arrival and 30-day mortality. Given the dichotomous nature of these variables, the models will be developed with logistic regression. The models will be first developed when the study will enroll the minimum number of necessary patients, and periodically updated. In this way, predictive tools capable of making accurate estimates of the prognosis of patients will be returned to the clinical practice as soon as possible.
The study will record the use of treatments and drugs, such as antivirals, ACE inhibitors and sartans, whose effect on COVID-19 patients is not yet clear and mostly based on small studies. The large cohort of patients will be used to perform CER analyses and evaluate the effect of these treatments on the prognosis of patients, with appropriate statistical methodology to account for the observational nature of the data. These results will be used to refine and improve the experimental design of this project.
